1. Field of the Invention
The present invention relates to a fuel injection control device that is applied to a fuel supply system injecting fuel through a predetermined injector and that controls a fuel injection characteristic of the system.
2. Description of Related Art
As one of technologies for improving emission of an in-vehicle diesel engine or the like, there is a known technology (high-pressure injection) for injecting fuel, which is pressurized to high pressure, directly into a cylinder through an injector. In recent years, a common rail fuel injection system (for example, a system described in Patent document 1: JP-A-H10-220272) has come to attract attention as a fuel injection system realizing the high-pressure injection. In the system, the fuel pumped from a fuel pump is accumulated in a common rail at a high-pressure state, and the accumulated high-pressure fuel is supplied to the injectors of respective cylinders through pipes (high-pressure fuel passages) provided to the respective cylinders. In the system, a predetermined pressure sensor (a rail pressure sensor) is provided to the common rail. The system is configured to control drive of various devices constituting a fuel supply system based on an output (a sensor output) of the rail pressure sensor.
Conventionally, in the case where an injection operation of the injector is controlled with such the common rail fuel injection system, a control method of setting an injection pattern in accordance with an engine operation state of each time with reference to a map (an adaptation map), in which an injection pattern (i.e., an adaptation value) for each engine operation state is written, or a mathematical expression is widely adopted. The device stores the optimum pattern (i.e., the adaptation value), which is beforehand obtained for each anticipated engine operation state through experiment and the like, as the map, the mathematical expression or the like (in ROM, for example). Thus, the device sets the injection pattern corresponding to the engine operation state with reference to the map, the mathematical expression or the like.
However, when mass production and mass marketing of respective components of the engine control system are performed, usually, there occur certain individual differences in characteristics of various kinds of control components including the injector, for example, between the engines and also between the cylinders in the case of a multi-cylinder engine. In such the case, it takes a lot of works and is not realistic for the present production system to obtain the adaptation values (the optimum injection patterns) for all the components (for example, all the cylinders manufactured through the mass production and mounted in the vehicle) in consideration of also the individual differences. Therefore, it is difficult to perform the control in consideration of all the influences due to the individual differences even when the map, in which the adaptation values are written, or the mathematical expression is used.
In order to perform the injection control with high accuracy, a characteristic change resulting from aging of the control components and the like is not negligible. Even if the conventional device such as the device described in Patent document 1 can obtain the optimum value with high accuracy in an initial stage, an influence of the subsequent characteristic change is unknowable. Therefore, there is a concern that there occurs a deviation from the optimum value with passage of time. In this case, an adaptation value of a degradation factor (a coefficient concerning a degree of degradation with time) may be beforehand obtained with an experimental value and the like and may be stored as a map, a mathematical expression or the like. However, there is also the above-mentioned individual difference in the temporal characteristic change of every component. Therefore, it is difficult to thoroughly remove the influence.